The present disclosure relates to an image processing apparatus and an image processing method.
For example, devices have become widely available which are capable of allowing a user to recognize displayed images as a stereoscopic image, by displaying an image corresponding to the left eye of the user (hereinafter, expressed as “left eye image” or “L image”) and an image corresponding to the right eye of the user (hereinafter, expressed as “right eye image” or “R image”) on a display screen. Devices such as those described above allow a user to recognize displayed images as a stereoscopic image by using parallax.
Further, display devices have been developed which are capable of displaying images with a plural number of viewpoints, such as 6 viewpoints or 9 viewpoints. In devices such as those described above, an image with a plural amount of viewpoints is displayed, at light condensing distances in the image corresponding to each of the viewpoints (hereinafter, there will be cases where these are expressed as “viewpoint images”), by arranging a plurality of viewpoint images with a different parallax, and periodically repeating the display of the viewpoint images horizontally with respect to the display screen.
Therefore, for example, since a plurality of normal vision regions are included in display devices such as those described above, having a plurality of users (viewers who view an image displayed on the display screen. Hereinafter, this will be assumed to be the same) simultaneously view a stereoscopic image will be realized. Further, for example, since display devices such as those described above are capable of allowing different viewpoint images to be viewed in accordance with the positions at which the users view the image on the display screen (hereinafter, there will be cases where this is expressed as “viewing position”), a stereoscopic image in which motion parallax reappears can be provided to the users.
Here, for example, images captured at different positions by an imaging apparatus can be included as viewpoint images corresponding to each of a plurality of viewpoints. However, in the case where viewpoint images corresponding to each of a plurality of viewpoints are set as images captured at different positions by an imaging apparatus, it will be costly. Accordingly, for example, technology has been developed related to the generation of images in virtual viewpoint positions (images in virtual imaging positions), based on a left eye image and a right eye image which constitute a stereoscopic image. For example, technology is included in JP 2006-115198A as technology related to the generation of images in virtual viewpoint positions, for example, based on a left eye image and a right eye image which constitute a stereoscopic image.